The number of veterans with Alzheimer?s disease (AlzD) is dramatically increasing. Testosterone is hypothesized to have an important role in risk, onset, and progression of AlzD in men. Testosterone deficiency prior to AlzD onset can promote AlzD pathophysiology during the preclinical phase when interventions may have the greatest success. Androgen deprivation therapy (ADT) for prostate cancer and testosterone replacement therapy (TRT) for age-related hypogonadism that directly target testosterone levels provide valuable opportunities to examine the impact of this androgen on the risk of developing AlzD, its prodromal condition mild cognitive impairment (MCI), and co-morbid health conditions. In the proposed study, we will utilize genetic and clinical data from the Million Veteran Program (MVP) to determine the impact of low testosterone, ADT, and TRT on the relative risk for MCI and AlzD. We hypothesize that the long-term impact of low testosterone, ADT, and TRT will be regulated by polygenic risk for AlzD and variation in the androgen receptor (AR) gene. These genetic factors represent critical, pre-existing markers of disease vulnerability and hormone sensitivity that have yet to be incorporated into research on protective and beneficial effects of testosterone on cognition and health. In our study, we will first determine if the risks of MCI, AlzD, and co- morbid medical disorders that promote dementia are increased by low testosterone levels in the MVP cohort. Next, we will determine if the clinical manipulation of testosterone levels is associated with differences in risks for MCI and AlzD. We hypothesize that individuals who undergo ADT will be at increased risk for MCI, AlzD and co-morbid health conditions relative to men with a history of prostate cancer but who have not been treated with ADT. We further hypothesize that men who receive TRT will be at decreased risks for these same conditions relative to men with low testosterone who do not receive hormone replacement therapy. We will next test whether pre-existing genetic risk for AlzD modifies the impact of testosterone-centered treatments on MCI and AlzD. We hypothesize that the observed effects of ADT and TRT on the cognitive and health outcomes will be greater in individuals at greater genetic risk for AlzD. Finally, we will determine whether genetically driven changes in androgen sensitivity due to CAG repeat length or ligand and DNA binding domain mutations regulate the long-term risks and benefits of ADT and TRT. We hypothesize that a more sensitive variant of the AR gene will confer greater risk for MCI and AlzD following ADT. Furthermore, we hypothesize that more androgen-sensitive individuals will benefit more from TRT, and will therefore show a reduced risk for MCI and AlzD. With its large sample size, extensive genetic data, and curated medical record data, the MVP represents an ideal resource with which to explore impact of testosterone-centered treatments on dementia risk, and provide a proof-of-principle model of genetically informed precision medicine. By evaluating the long- term risks and benefits of ADT and TRT on cognitive health, in the context of a genetically informative design, the proposed study will provide valuable insights into the role of testosterone as a risk factor for MCI and AlzD, and increase our understanding of a critical AlzD-related pathway that may be responsive to intervention. The present study will also shed light on the long-term risks and benefits associated with ADT and TRT, and clarify whether these treatments, which are efficacious in certain contexts, set in motion or delay pathophysiological processes that result in the emergence of AlzD symptoms in men. Finally, we will determine if our hypothesis driven, genetically informed study investigating underlying genetic risk (i.e., the polygenic risk for AlzD) and genetically determined treatment sensitivity (i.e., variation within the AR gene) can improve treatment efficacy and evidence-based usage of TRT and reduce detrimental outcomes of ADT. We believe our study is ideal for the MVP, and will demonstrate effectiveness of the broader program for biomedical research and the advancement of personalized/precision medicine.